*THE NATIONAL COLLABORATIVE STUDY OF EARLY PSYCHOSIS AND SUICIDE (NCSEPS) The National Collaborative Study of Early Psychosis and Suicide (NCSEPS), is conducted in conjunction with the Department of Defense (DoD) and the Department of Veterans Affairs (VA). The NCSEPS population is active duty personnel hospitalized for either bipolar disorder, major depressive disorder, or schizophrenia while in the military, as well as a set of matched controls. The goals of the NCSEPS are: 1) to assess the characteristics of this first-episode population; 2) to determine rates of illness of these disorders in this population; 3) to detect and decrease the occurrence of long-term morbidity associated with these disorders; 4) to determine what benefits result from early intervention in patients first manifesting signs of these disorders; and 5) to decrease the risk of suicide in this population. The Armed Forces provide an excellent opportunity to study the potential benefits of early intervention in the major psychoses. Historically, a number of articles published over the course of this century indicate that a high percentage of individuals who become psychotic while in the military do so early in the course of their military career. Furthermore, there are a number of potential advantages to carrying out such a project within the DoD: the highest risk period for developing schizophrenia and bipolar disorder occurs in the age range when most individuals enter the Armed Forces; the controlled environment individuals experience in the Armed Forces provides an almost unique opportunity to provide help; and studying servicemen and servicewomen provides the potential to follow-up those individuals who develop an illness through the VA. To date, NCSEPS has amassed sociodemographic characteristics and hospitalization data on approximately 13,000 patients and over 135,000 controls. Although analyses of these data are ongoing, a number of interesting preliminary findings have emerged. HEIGHT, WEIGHT, BMI, AND DIAGNOSIS. In a paper published this year, NCSEPS researchers investigated the idea that individuals with certain body types were prone to particular psychiatric diseases. The paper investigated the relationship between height, weight, and BMI and psychiatric diagnoses, and found no differences in height, weight, or BMI between patients (N=7500) with bipolar disorder, major depressive disorder, or schizophrenia and a large comparison group (N=86,000). USE OF VA SERVICES BY NCSEPS SAMPLE. In a paper published this year, NCSEPS researchers used demographic and diagnostic data to examine patterns of use of VA mental health outpatient services in over 2500 individuals hospitalized with bipolar disorder, major depressive disorder, or schizophrenia while on active duty and then subsequently discharged from the military. Patients were followed for at least two years and were all eligible for VA services. A little over half the sample (52%) received outpatient care. Individuals with schizophrenia and bipolar disorder, women, and older individuals were all more likely to seek VA care. INCIDENCE OF BIPOLAR DISORDER, MAJOR DEPRESSIVE DISORDER, AND SCHIZOPHRENIA IN ACTIVE DUTY MILITARY PERSONNEL. In a paper submitted for publication, the incidence of bipolar disorder (0.20 per 1000), major depressive disorder (0.72 per 1000), and schizophrenia (0.16 per 1000) was ascertained in active duty personnel hospitalized for the first time with these disorders. Incidence of bipolar disorder and major depressive disorder were greater in females than in males, but there were no gender differences for schizophrenia. Blacks had a higher rate of schizophrenia than whites. Because of the large size of the NCSEPS study group (8700 patients), these data are an important addition to existing information on psychiatric incidence rates. THE IMPACT OF LIFE EVENTS ON THE DEVELOPMENT OF PSYCHIATRIC DISORDERS. With collaborators at Walter Reed Army Medical Center (WRAMC), NCSEPS researchers reviewed medical boards for individuals (N=74) hospitalized at WRAMC who became mentally ill within their first year of military service. Almost 80% had a new (vs. a pre-existing) psychosis. Onset of illness occurred on average at about 180 days for most patients. Precipitant stressors at particular points in time were not apparent for these patients, suggesting life events may be more important for relapse than onset. These results have been submitted for publication. RESULTS FROM THE NATIONAL DEATH INDEX (NDI). In a paper currently being prepared for publication, mortality was ascertained in over 7500 individuals diagnosed with bipolar disorder, major depressive disorder, or schizophrenia while in the U.S. military who were subsequently discharged. Rates of death from all causes excluding suicide were not elevated compared with the general population. SMRs for bipolar (9.7) and major depressive disorder (8.2) were lower than those reported by most studies of psychiatric patients; suicide SMRs for individuals with schizophrenia (23.6) were higher. The findings suggest that the transition from military to civilian life is difficult for all patients with psychotic disorders, but particularly for those with schizophrenia, and that additional resources should be directed toward protecting this group from suicide. Mortality data for psychiatrically healthy controls are currently being ascertained. HOSPITALIZATION DURING THE FIRST 15 DAYS OF ACTIVE DUTY. The percentage of patients hospitalized within the first year of active duty who were also hospitalized within the first 15 days was 40.1% of those with bipolar disorder, 14.3% percent of those with major depressive disorder, and 37.3% percent of those with schizophrenia. All three patient groups had a high risk of hospitalization during the first 15 days of active duty, although the risk was much greater for bipolar disorder and schizophrenia. SOCIOECONOMIC STATUS (SES) FOR NCSEPS PATIENTS. In the NCSEPS study group, blacks received a diagnosis of schizophrenia more often than whites. This ethnic difference has been found in other studies of schizophrenia. To examine whether SES might explain the higher than expected rate, median household income from the 1990 US Census for each recruit?s home zip code was determined. Black schizophrenics were found to be from the poorest subgroup in the NCSEPS population, but the median income for all blacks surveyed by the 1990 Census was considerably lower. This suggests that another factor besides SES is at work. The NCSEPS project is particularly important because of its ability to follow a large number of patients before the onset of their illness through the VA system, and because of the availability of substantial amounts of pre- and post-morbid data.